Search

Your search keyword '"Kidiyo Kpalma"' showing total 177 results
Start Over Author "Kidiyo Kpalma"
177 results on '"Kidiyo Kpalma"'

65. An Overview of Advances of Pattern Recognition Systems in Computer Vision

Author

Kidiyo Kpalma, Joseph Ronsin, Institut d'Electronique et de Télécommunications de Rennes (IETR), Centre National de la Recherche Scientifique (CNRS)-Ecole Supérieure d'Electricité - SUPELEC (FRANCE)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES), Prof. Goro Obinata and Dr. Ashish Dutta, Kpalma, Kidiyo, Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), and Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Ecole Supérieure d'Electricité - SUPELEC (FRANCE)-Centre National de la Recherche Scientifique (CNRS)

Subjects
Computer science, [INFO.INFO-TS] Computer Science [cs]/Signal and Image Processing, Class (philosophy), 02 engineering and technology, Similarity measure, Content-based image retrieval, Machine learning, computer.software_genre, image retrieval, Task (project management), curvature scale-space, shape description, content-based image retrieval, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, pattern recognition system, 0202 electrical engineering, electronic engineering, information engineering, Feature (machine learning), Computer vision, Image retrieval, IPM function, [SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing, shape matching, business.industry, 020207 software engineering, Pattern recognition, Object (computer science), similarity measure, features extraction, Pattern recognition (psychology), 020201 artificial intelligence & image processing, Artificial intelligence, business, computer, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing
Abstract

As mentioned before, pattern recognition does not appear as a new problem. A lot of studies have been performed on this scientific field and a lot of works are currently developed. Pattern recognition is a wide topic in machine learning. It aims to classify a pattern into one of a number of classes. It appears in various fields like psychology, agriculture, computer vision, robotics , biometrics... With technological improvements and growing performances of computer science, its application field has no real limitation. In this context, a challenge consists of finding some suitable description features since commonly, the pattern to be classified must be represented by a set of features characterising it. These features must have discriminative properties: efficient features must be affined transformations insensitive. They must be robust against noise and against elastic deformations due, e.g., to movement in pictures. Through the application example based on our MSGPR method, we have illustrated various aspects of a PRS. With this example, we have illustrated the description task that enabled us to extract multi-scale features from the generated IPM function. By using theses features in the classification task, we identified the letters from a car number plate so that we automatically retrieved the license number of a vehicle.

Published
2021

66. Video-based Facial Micro-Expression Analysis: A Survey of Datasets, Features and Algorithms

Author

Xianye Ben, Yi Ren, Kidiyo Kpalma, Yong-Jin Liu, Junping Zhang, Weixiao Meng, Su-Jing Wang, Shandong University, Fudan University [Shanghai], Chinese Academy of Sciences [Beijing] (CAS), Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Harbin Institute of Technology (HIT), Tsinghua University [Beijing] (THU), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Nantes Université (NU)-Université de Rennes 1 (UR1)

Subjects
FOS: Computer and information sciences, Computer science, Computer Vision and Pattern Recognition (cs.CV), Emotions, Computer Science - Computer Vision and Pattern Recognition, 02 engineering and technology, Facial recognition system, [SPI]Engineering Sciences [physics], Artificial Intelligence, 0202 electrical engineering, electronic engineering, information engineering, Humans, Transient (computer programming), Macro, Video based, Structure (mathematical logic), Facial expression, Applied Mathematics, Spotting, Facial Expression, Computational Theory and Mathematics, Key (cryptography), 020201 artificial intelligence & image processing, Computer Vision and Pattern Recognition, Algorithm, Software, Algorithms
Abstract

International audience; Unlike the conventional facial expressions, micro-expressions are involuntary and transient facial expressions capable of revealing the genuine emotions that people attempt to hide. Therefore, they can provide important information in a broad range of applications such as lie detection, criminal detection, etc. Since micro-expressions are transient and of low intensity, however, their detection and recognition is difficult and relies heavily on expert experiences. Due to its intrinsic particularity and complexity, video-based micro-expression analysis is attractive but challenging, and has recently become an active area of research. Although there have been numerous developments in this area, thus far there has been no comprehensive survey that provides researchers with a systematic overview of these developments with a unified evaluation. Accordingly, in this survey paper, we first highlight the key differences between macro- and micro-expressions, then use these differences to guide our research survey of video-based micro-expression analysis in a cascaded structure, encompassing the neuropsychological basis, datasets, features, spotting algorithms, recognition algorithms, applications and evaluation of state-of-the-art approaches. For each aspect, the basic techniques, advanced developments and major challenges are addressed and discussed. Furthermore, after considering the limitations of existing micro-expression datasets, we present and release a new dataset - called micro-and-macro expression warehouse (MMEW) - containing more video samples and more labeled emotion types. We then perform a unified comparison of representative methods on CAS(ME) for spotting, and on MMEW and SAMM for recognition, respectively. Finally, some potential future research directions are explored and outlined.

Published
2021
Full Text
View/download PDF

70. Geometric Distance for Fast Micro-Expression Detection

Author

Mingqiang Yang, Kidiyo Kpalma, Dengwang Li, Hua Lu, Shandong Normal University, Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Shandong University, Natural Science Foundation of Shandong Province: ZR2019PF011201902028023, Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Computer science, business.industry, Feature extraction, 020207 software engineering, Geometric distance, Pattern recognition, 02 engineering and technology, Facial recognition system, Expression (mathematics), Euclidean distance, Set (abstract data type), [SPI]Engineering Sciences [physics], 0202 electrical engineering, electronic engineering, information engineering, micro-expression detection, 020201 artificial intelligence & image processing, support vector machine, Artificial intelligence, business
Abstract

International audience; In this paper, we propose a new model for micro-expression detection in videos in conjunction with a set of facial keypoints. The main contribution lies at the construction of geometric features extracted using the geometric distances between pairs of keypoints in different groups. The proposed geometric features-based micro-expression detection model requires a very low computation complexity and simultaneously attains highly accurate detection rates. We compare the proposed geometric features-based model to the existing state-of-the-art micro-expression detection and recognition models in three datasets, where the proposed method has achieved better, or at least comparably accurate results but requiring much less computation time. © 2020 IEEE.

Published
2020
Full Text
View/download PDF

71. Review of Recent Deep Learning Based Methods for Image-Text Retrieval

Author

Jianan Chen, Kidiyo Kpalma, Lu Zhang, Cong Bai, Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA), Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Zhejiang University of Technology, Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Information retrieval, Modalities, business.industry, Computer science, Deep learning, 020206 networking & telecommunications, Context (language use), 02 engineering and technology, Popularity, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], [INFO.INFO-IR]Computer Science [cs]/Information Retrieval [cs.IR], 0202 electrical engineering, electronic engineering, information engineering, Benchmark (computing), Key (cryptography), 020201 artificial intelligence & image processing, Social media, Artificial intelligence, business, Mobile device
Abstract

International audience; Cross-modal retrieval has drawn much attention in recent years due to the diversity and the quantity of information data that exploded with the popularity of mobile devices and social media. Extracting relevant information efficiently from large-scale multi-modal data is becoming a crucial problem of information retrieval. Cross-modal retrieval aims to retrieve relevant information across different modalities. In this paper, we highlight key points of recent cross-modal retrieval approaches based on deep-learning, especially in the image-text retrieval context, and classify them into four categories according to different embedding methods. Evaluations of state-of-the-art cross-modal retrieval methods on two benchmark datasets are shown at the end of this paper.

Published
2020
Full Text
View/download PDF

72. A semantics-guided warping for semi-supervised video object instance segmentation

Author

Kidiyo Kpalma, Qiong Wang, Lu Zhang, Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), China Scholarship Council, CSC: 201504490048National Basic Research Program of China (973 Program): 2018YFE0126100, Campilho A.Karray F.Wang Z., Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Warping flow, Computer science, business.industry, Computation, Frame (networking), Process (computing), ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, 02 engineering and technology, Semantics, Object (computer science), Domain (software engineering), [SPI]Engineering Sciences [physics], 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Computer vision, Segmentation, Artificial intelligence, Semi-supervised video object instance segmentation, Image warping, business
Abstract

International audience; Unlabelled - Medicinally active compounds in the flavonoid class of phytochemicals are being studied for antiviral action against various DNA and RNA viruses. Quercetin is a flavonoid present in a wide range of foods, including fruits and vegetables. It is said to be efficient against a wide range of viruses. This research investigated the usefulness of Quercetin against Hepatitis C virus, Dengue type 2 virus, Ebola virus, and Influenza A using computational models. A molecular docking study using the online tool PockDrug was accomplished to identify the best binding sites between Quercetin and PubChem-based receptors. Network-pharmacological assay to opt to verify function-specific gene-compound interactions using STITCH, STRING, GSEA, Cytoscape plugin cytoHubba. Quercetin explored tremendous binding affinity against NS5A protein for HCV with a docking score of - 6.268 kcal/mol, NS5 for DENV-2 with a docking score of - 5.393 kcal/mol, VP35 protein for EBOV with a docking score of - 4.524 kcal/mol, and NP protein for IAV with a docking score of - 6.954 kcal/mol. In the network-pharmacology study, out of 39 hub genes, 38 genes have been found to interact with Quercetin and the top interconnected nodes in the protein-protein network were (based on the degree of interaction with other nodes) and Negative binding energies were noticed in Quercetin-receptor interaction. Results demonstrate that Quercetin could be a potential antiviral agent against these viral diseases with further study in models. Supplementary information - The online version contains supplementary material available at 10.1007/s40203-022-00132-2.

Published
2020
Full Text
View/download PDF

79. A PCA-PD fusion method for change detection in remote sensing multi temporal images

Author

Nasreddine Taleb, Soltana Achour, Kidiyo Kpalma, Joseph Ronsin, Miloud Chikr El-Mezouar, Université Djilali Liabès [Sidi-Bel-Abbès], Institut d'Électronique et des Technologies du numéRique (IETR), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Fusion, 010504 meteorology & atmospheric sciences, Computer science, Geography, Planning and Development, 0211 other engineering and technologies, Image processing, 02 engineering and technology, 01 natural sciences, Remote sensing (archaeology), [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Change detection, ComputingMilieux_MISCELLANEOUS, 021101 geological & geomatics engineering, 0105 earth and related environmental sciences, Water Science and Technology, Remote sensing
Abstract

In remote sensing, for applications as environment monitoring, change detection based on image processing is one of the most important techniques. To reach high performance various techniques of fu...

Published
2020
Full Text
View/download PDF

80. Local Binary Pattern and Its Variants: Application to Face Analysis

Author

Kidiyo Kpalma, Hua Lu, Joseph Ronsin, Vincent Débordès, Jade Lizé, Institut d'Électronique et des Technologies du numéRique (IETR), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), School of Physics and Electronics, Shandong Normal University, Institut d'Electronique et de Télécommunications de Rennes (IETR), Centre National de la Recherche Scientifique (CNRS)-Ecole Supérieure d'Electricité - SUPELEC (FRANCE)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Ecole Supérieure d'Electricité - SUPELEC (FRANCE)-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Pathogen-associated molecular patterns (PAMPs), Facial expression, business.industry, Local binary patterns, Computer science, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Face analysis, 020207 software engineering, Pattern recognition, 02 engineering and technology, Tomato resistance to virus, Hrip1, Facial recognition system, Local structure, Set (abstract data type), Tomato yellow leaf curl virus (TYLCV), ComputingMethodologies_PATTERNRECOGNITION, Transcriptional and posttranscriptional regulation, 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Artificial intelligence, business, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing
Abstract

Source : First international Conference on Smart Information & Communication Technologies (SmartICT’19), 26-09-2019, Saidia, Maroc.; International audience; Unlabelled - Tomato yellow leaf curl virus (TYLCV) causes tremendous losses of tomato worldwide. An elicitor Hrip1, which produced by , can serve as a pathogen-associated molecular patterns (PAMPs) to trigger the immune defense response in . Here, we show that Hrip1 can be targeted to the extracellular space and significantly delayed the development of symptoms caused by TYLCV in tomato. In basis of RNA-seq profiling, we find that 1621 differential expression genes (DEGs) with the opposite expression patterns are enriched in plant response to biotic stress between Hrip1 treatment and TYLCV infection of tomato. Thirty-two known differential expression miRNAs with the opposite expression patterns are identified by small RNA sequencing and the target genes of these miRNAs are significantly enriched in phenylpropanoid biosynthesis, plant hormone signal transduction and peroxisome. Based on the Pearson correlation analysis, 13 negative and 21 positive correlations are observed between differential expression miRNAs and DEGs. These miRNAs, which act as a key mediator of tomato resistance to TYLCV induced by Hrip1, regulate the expression of phenylpropanoid biosynthesis and plant hormone signal transduction-related genes. Taken together, our results provide an insight into tomato resistance to TYLCV induced by PAMP at transcriptional and posttranscriptional levels. Supplementary information - The online version contains supplementary material available at 10.1007/s13205-022-03426-6.

Published
2020
Full Text
View/download PDF

81. Overview of deep-learning based methods for salient object detection in videos

Author

Yan Li, Kidiyo Kpalma, Qiong Wang, Lu Zhang, Zhejiang University of Technology, Institut d'Électronique et des Technologies du numéRique (IETR), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Université libre de Bruxelles (ULB), China Scholarship Council (CSC)China Scholarship Council [201504490048], National Key Research and Development Program of China [2018YFE0126100], Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Salient object detection, business.industry, Computer science, Deep learning, Video, 02 engineering and technology, Machine learning, computer.software_genre, Deep-learning, 01 natural sciences, Domain (software engineering), [SPI]Engineering Sciences [physics], Artificial Intelligence, 0103 physical sciences, Signal Processing, 0202 electrical engineering, electronic engineering, information engineering, Benchmark (computing), 020201 artificial intelligence & image processing, Computer Vision and Pattern Recognition, Artificial intelligence, 010306 general physics, business, computer, Software
Abstract

International audience; Video salient object detection is a challenging and important problem in computer vision domain. In recent years, deep-learning based methods have contributed to significant improvements in this domain. This paper provides an overview of recent developments in this domain and compares the corresponding methods up to date, including 1) Classification of the state-of-the-art methods and their frameworks; 2) summary of the benchmark datasets and commonly used evaluation metrics; 3) experimental comparison of the performances of the state-of-the-art methods; 4) suggestions of some promising future works for unsolved challenges.

Published
2020
Full Text
View/download PDF

82. Reducing lbp features for facial identification and expression recognition

Author

Kidiyo Kpalma, Joseph Ronsin, Hua Lu, Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Shandong Normal University, El Moussati A.Kpalma K.Ghaouth Belkasmi M.Saber M.Guegan S., Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Pixel, Computer science, Local binary patterns, business.industry, Texture Descriptor, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Byte, Pattern recognition, 02 engineering and technology, 01 natural sciences, Facial identification, Textural approach, Reduction (complexity), Identification (information), Macro-expression recognition, [SPI]Engineering Sciences [physics], Histogram, LBP, 0103 physical sciences, 0202 electrical engineering, electronic engineering, information engineering, Code (cryptography), 020201 artificial intelligence & image processing, Artificial intelligence, 010306 general physics, business
Abstract

International audience; The LBP (Local Binary Pattern) texture descriptor has demonstrated its superiority in several image applications texture characterization, facial identification and macro-expression recognition. Featuring by LBP characterizes image by its local structures, observing its micro patterns and building a histogram. Each observed pixel is featured and then encoded into one byte. All these codes constitute bins for the histogram. For an efficient classification, encoded bytes can be divided into uniform and non-uniform codes. In standard applications, only uniform codes are used leading to 59 codes. The present work proposes an additional process for the reduction of these codes. The proposal is developed and comparatively evaluated with success to classical LBP. Experimental evaluations are performed on 2 different databases for facial identification and macro-expression recognition respectively, and this for different reduction of code length. Though for macro-expression recognition the proposed features can give lower but comparable performance with the traditional LBP, for facial identification they perform very well and keep excellent efficiency. This approach can be extended to most part of LBP variants while keeping the simplicity of LBP. © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2020.

Published
2019
Full Text
View/download PDF

83. A SPATIOTEMPORAL DEEP LEARNING SOLUTION FOR AUTOMATIC MICRO-EXPRESSIONS RECOGNITION FROM LOCAL FACIAL REGIONS

Author

Amel Benazza-Benyahia, Mouath Aouayeb, Kidiyo Kpalma, Wassim Hamidouche, Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Unité de Recherche en Imagerie Satellitaire et ses Applications (URISA), SUP'COM, Université de Nantes (UN)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Nantes Université (NU)-Université de Rennes 1 (UR1)

Subjects
Index Terms-Micro-Expression, Computer science, business.industry, Deep learning, Regions of In- terest, Feature extraction, 020206 networking & telecommunications, Pattern recognition, 02 engineering and technology, Convolutional neural network, Convolution, Support vector machine, Long short term memory, Face (geometry), [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV], 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Artificial intelligence, State (computer science), business, LSTM, CNN
Abstract

International audience; Humans always try to hide their Macro-Expressions (MaE) to conceal their real emotion, and it is hard to distinguish between true and false emotions even with artificial intelligence. Micro-Expressions (MiEs), on the contrary, are spontaneous and fast, undetectable with the naked eye and thus always inform us of true feelings. Therefore , there is plenty of studies to generate an automatic system of detecting and analyzing these MiEs. In this paper we propose a new solution that relies on a combination of Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM) applied on particular regions of the face to extract relevant spatial and temporal features, respectively, for MiEs recognition. The proposed solution achieves high recognition accuracy of 90% precision on a different databases including SMIC, CASME II and SAMM. Moreover, under the conditions of Micro-Expression Grand Challenge (MEGC) 2019, our approach performs better than the state of the art solutions including the ones proposed in the challenge.

Published
2019

84. Advances in Smart Technologies Applications and Case Studies : Selected Papers From the First International Conference on Smart Information and Communication Technologies, SmartICT 2019, September 26-28, 2019, Saidia, Morocco

Author

Ali El Moussati, Kidiyo Kpalma, Mohammed Ghaouth Belkasmi, Mohammed Saber, Sylvain Guégan, Ali El Moussati, Kidiyo Kpalma, Mohammed Ghaouth Belkasmi, Mohammed Saber, and Sylvain Guégan

Subjects
Wireless communication systems, Mobile communication systems, Signal processing, Information technology—Management
Abstract

This book offers a comprehensive snapshot of practically-relevant developments in the field of smart information technologies. Including selected papers presented at the First International Conference on Smart Information and Communication Technologies, SmartICT 2019, held on September 26-28, in 2019, Saidia, Morocco, it reports on practical findings, and includes useful tutorials concerning current technologies and suggestions to improve them. It covers a wide range of applications, from health and energy management, to digital education, agriculture and cybersecurity, providing readers with a source of new ideas for future research and collaborations.

Published
2020

86. Motion descriptors for micro-expression recognition

Author

Kidiyo Kpalma, Joseph Ronsin, Hua Lu, Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Computer science, Optical flow, 02 engineering and technology, Motion (physics), MBH, [INFO.INFO-NI]Computer Science [cs]/Networking and Internet Architecture [cs.NI], HOF, Histogram, 0202 electrical engineering, electronic engineering, information engineering, Feature (machine learning), Electrical and Electronic Engineering, Differential (infinitesimal), Protocol (object-oriented programming), Series (mathematics), business.industry, FMBH, 020206 networking & telecommunications, Pattern recognition, [SPI.TRON]Engineering Sciences [physics]/Electronics, Micro-expression recognition, Facial expression recognition, Signal Processing, 020201 artificial intelligence & image processing, Computer Vision and Pattern Recognition, Artificial intelligence, business, Software
Abstract

International audience; Micro-expression analysis is an interesting and challenging task in computer vision. It has inspired a series of possible applications in many areas such as police–criminal interrogation and important business negotiation. One of the most crucial step in a micro-expression recognition system is the extraction of well discriminating features. In this paper, we propose a new feature based on the fusion of motion boundary histograms (FMBH). This feature is generated by combining both the horizontal and the vertical components of the differential of optical flow as inspired from the motion boundary histograms (MBH). The proposed feature is then validated and evaluated through the leave-one-subject-out (LOSO) protocol for micro-expression recognition. Moreover, the proposed method is compared to state-of-the-art methods on four well-known databases CASME, CASME II, SMIC and CAS(ME)2. Comparative experimental results demonstrate that the proposed FMBH feature descriptor yields promising performance. © 2018 Elsevier B.V.

Published
2018
Full Text
View/download PDF

87. Best-Performing Color Space for Land-Sea Segmentation

Author

Amadou Seidou Maiga, Oumar Diop, Kidiyo Kpalma, Seynabou Toure, Université Gaston Berger de Saint-Louis Sénégal (UGB), Institut d'Électronique et des Technologies du numéRique (IETR), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Herencsar N. (ed), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Land-sea segmentation, Computer science, business.industry, feature extraction, FOOS, Feature extraction, 0211 other engineering and technologies, k-means clustering, YCbCr, 02 engineering and technology, HSL and HSV, color space, Space (commercial competition), Color space, [SPI]Engineering Sciences [physics], 0202 electrical engineering, electronic engineering, information engineering, RGB color model, 020201 artificial intelligence & image processing, Computer vision, Segmentation, Artificial intelligence, business, K-means, 021101 geological & geomatics engineering
Abstract

International audience; Land-sea segmentation is one of the most challenging tasks in remote sensing. It is of great importance for many applications like coastline extraction and near shore ship detection. To facilitate future land-sea segmentation related research, we present, in this paper, an experimental study conducted on five color spaces (RGB, XYZ, La∗b∗, HSV and YCbCr) in order to show the best-performing color space for land-sea segmentation. The test is carried out on 50 images. The average segmentation results are given. The results show that the luminance-chrominance space (YCbCr, La∗b∗, HSV) are more appropriated than primary space (RGB, XYZ), in general. It is particularly noted that the perceptual color space HSV allows better results in land-sea segmentation application. © 2018 IEEE.

Published
2018
Full Text
View/download PDF

88. Facial Identification and Macro Expression Recognition with a New Textural Featuring Approach

Author

Kidiyo Kpalma, Joseph Ronsin, Hua Lu, Mingqiang Yang, Institut d'Electronique et de Télécommunications de Rennes (IETR), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Ecole Supérieure d'Electricité - SUPELEC (FRANCE)-Centre National de la Recherche Scientifique (CNRS), Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Shandong University, Centre National de la Recherche Scientifique (CNRS)-Ecole Supérieure d'Electricité - SUPELEC (FRANCE)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Facial expression recognition, Computer science, business.industry, Pattern recognition, Identification (biology), Artificial intelligence, Macro, business, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, ComputingMilieux_MISCELLANEOUS
Abstract

International audience

Published
2018
Full Text
View/download PDF

89. COASTLINE DETECTION USING FUSION OF OVER SEGMENTATION AND DISTANCE REGULARIZATION LEVEL SET EVOLUTION

Author

Oumar Diop, Kidiyo Kpalma, Amadou Seidou Maiga, Seynabou Toure, Institut d'Électronique et des Technologies du numéRique (IETR), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Ground truth, Level set (data structures), Computer science, Interface (Java), business.industry, 0211 other engineering and technologies, k-means clustering, Wavelet transform, Pattern recognition, 02 engineering and technology, Regularization (mathematics), 0202 electrical engineering, electronic engineering, information engineering, Littoral zone, 020201 artificial intelligence & image processing, 14. Life underwater, Artificial intelligence, Limit (mathematics), business, Cluster analysis, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, ComputingMilieux_MISCELLANEOUS, 021101 geological & geomatics engineering
Abstract

Coastline detection is a very challenging task in optical remote sensing. However the majority of commonly used methods have been developed for low to medium resolution without specification of the key indicator that is used. In this paper, we propose a new approach for very high resolution images using a specific indicator. First, a pre-processing step is carried out to convert images into the optimal colour space (HSV). Then, wavelet decomposition is used to extract different colour and texture features. These colour and texture features are then used for Fusion of Over Segmentation (FOOS) based clustering to have the distinctive natural classes of the littoral. Among these classes are waves, dry sand, wet sand, sea and land. We choose the mean level of high tide water, the interface between dry sand and wet sand, as a coastline indicator. To find this limit, we use a Distance Regularization Level Set Evolution (DRLSE), which automatically evolves towards the desired sea-land border. The result obtained is then compared with a ground truth. Experimental results prove that the proposed method is an efficient coastline detection process in terms of quantitative and visual performances.

Published
2018

90. A New Approach to Region Based Image Retrieval using Shape Adaptive Discrete Wavelet Transform

Author

Kidiyo Kpalma, Kamel Belloulata, Lakhdar Belhallouche, Université Djilali Liabès [Sidi-Bel-Abbès], Institut d'Électronique et des Technologies du numéRique (IETR), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Université de Nantes (UN)-Université de Rennes 1 (UR1), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), and Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)

Subjects
Region-based image retrieval (RBIR), Discrete wavelet transform, business.industry, Computer science, Second-generation wavelet transform, Stationary wavelet transform, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Wavelet transform, SA-DWT, 020207 software engineering, Pattern recognition, 02 engineering and technology, Content-based image retrieval, Wavelet packet decomposition, Wavelet, Computer Science::Computer Vision and Pattern Recognition, Computer Science::Multimedia, Content-based image retrieval (CBIR), 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Artificial intelligence, DWT, business, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Image retrieval
Abstract

International audience; In this paper, we present an efficient region-based image retrieval method, which uses multi-features color, texture and edge descriptors. In contrast to recent image retrieval methods, which use discrete wavelet transform (DWT), we propose using shape adaptive discrete wavelet transform (SA-DWT). The advantage of this method is that the number of coefficients after transformation is identical to the number of pixels in the original region. Since image data is often stored in compressed formats: JPEG 2000, MPEG 4…; constructing image histograms directly in the compressed domain, allows accelerating the retrieval operation time, and reducing computing complexities. Moreover, SA-DWT represents the best way to exploit the coefficients characteristics, and properties such as the correlation. Characterizing image regions without any conversion or modification is first addressed. Using edge descriptor to complement image region characterizing is then introduced. Experimental results show that the proposed method outperforms content based image retrieval methods and recent region based image retrieval methods

Published
2016
Full Text
View/download PDF

91. Human Detection using HOG-SVM, Mixture of Gaussian and Background Contours Subtraction

Author

Kidiyo Kpalma, Abdourahman Houssein Ahmed, Abdoulkader Osman Guédi, Université de Djibouti, Institut d'Électronique et des Technologies du numéRique (IETR), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Computer science, Histogram of Oriented Gradients (HOG), Gaussian, 0211 other engineering and technologies, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Image processing, 02 engineering and technology, Silhouette, symbols.namesake, Histogram, 0202 electrical engineering, electronic engineering, information engineering, Segmentation, Computer vision, background contours subtraction, 021106 design practice & management, Background subtraction, business.industry, Subtraction, Object detection, mixture of Gaussian (MOG), support vector machines (SVM) classifier, symbols, 020201 artificial intelligence & image processing, Artificial intelligence, business, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Human detection
Abstract

International audience; Automatic moving object/Human detection in a video sequence is one of the most difficult problems in the field of image processing and computer vision. The HOG-SVM provides a detection windows that is not perfectly adjusted to the silhouette of the Human detected. It is possible to apply a postprocess based on background subtraction to improve the segmentation of the detection. In this paper, we present thus a detection method that improves results provided by HOG-SVM with a combination of mixture of Gaussian and background contours subtraction.

Published
2017
Full Text
View/download PDF

92. Fast filtering-based temporal saliency detection using Minimum Barrier Distance

Author

Kidiyo Kpalma, Lu Zhang, Qiong Wang, Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Pixel, Computer science, business.industry, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Optical flow, Boundary (topology), 020206 networking & telecommunications, 02 engineering and technology, Filter (signal processing), Object (computer science), Object detection, Motion (physics), Domain (software engineering), [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Computer vision, Artificial intelligence, business, ComputingMilieux_MISCELLANEOUS
Abstract

In the video salient object detection domain, the optical flow technique is widely used to extract temporal saliency. The big issue with this approach is that the global motion, when it exists, may be wrongly detected as a salient object. To cope with this problem, we propose to filter out the global motion by using the boundary connectivity cue. Firstly, an edge-aware filter called the guided filter is introduced to preprocess the color optical flow map for enhancing object edges. Then the pixel's connectivity to the boundary is achieved by using the minimum barrier distance in the filtered optical flow map, which leads to the effective removal of the global motion. The proposed method is assessed on the popular Seg-Track v2 and Fukuchi datasets and then compared to state-of-the-art methods. The experimental results show that the proposed approach outperforms the existing related methods.

Published
2017
Full Text
View/download PDF

93. Hierarchical Saliency Detection via Probabilistic Object Boundaries

Author

Kidiyo Kpalma, Haijun Lei, Hai Xie, Wenbin Zou, Nikos Komodakis, Xiaoli Sun, College of Computer Science and Software Engineering, Shenzhen University, College of Information Engineering, Shenzhen University, Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), College of Mathematics, Shenzhen University, École des Ponts ParisTech (ENPC), 61401287, NSFC, JCYJ20160506172651253, Natural Science Foundation of Shenzhen, 2016058, Natural Science Foundation of SZU, Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), and Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)

Subjects
Computer science, Bayesian probability, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, 02 engineering and technology, object segmentation, Artificial Intelligence, 0202 electrical engineering, electronic engineering, information engineering, Computer vision, 10. No inequality, Computational model, Saliency detection, business.industry, center bias, Probabilistic logic, 020207 software engineering, Pattern recognition, Object (computer science), Thresholding, Kadir–Brady saliency detector, Contour line, Computer Science::Computer Vision and Pattern Recognition, Benchmark (computing), 020201 artificial intelligence & image processing, Computer Vision and Pattern Recognition, Artificial intelligence, business, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Software
Abstract

International audience; Though there are many computational models proposed for saliency detection, few of them take object boundary information into account. This paper presents a hierarchical saliency detection model incorporating probabilistic object boundaries, which is based on the observation that salient objects are generally surrounded by explicit boundaries and show contrast with their surroundings. We perform adaptive thresholding operation on ultrametric contour map, which leads to hierarchical image segmentations, and compute the saliency map for each layer based on the proposed robust center bias, border bias, color dissimilarity and spatial coherence measures. After a linear weighted combination of multi-layer saliency maps, and Bayesian enhancement procedure, the final saliency map is obtained. Extensive experimental results on three challenging benchmark datasets demonstrate that the proposed model outperforms eight state-of-the-art saliency detection models.

Published
2017
Full Text
View/download PDF

94. Sparse representation based histogram in color texture retrieval

Author

Kidiyo Kpalma, Joseph Ronsin, Cong Bai, Jinglin Zhang, Jianan Chen, Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), 6164, CNRS, Conseil National de la Recherche Scientifique, LY15F020028, Natural Science Foundation of Zhejiang Province, LY15F030014, Natural Science Foundation of Zhejiang Province, LY16F020033, Natural Science Foundation of Zhejiang Province, S8113055001, NUIST, Nanjing University of Information Science and Technology, SBK2015040336, Jiangsu Province Natural Science Foundation, NSFC, National Science Foundation of China, Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Nantes Université (NU)-Université de Rennes 1 (UR1)

Subjects
Discrete wavelet transform, Color histogram, State-of-the-art approach, Feature vector, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Color, Wavelet decomposition, Feature vectors, 02 engineering and technology, Similarity measure, Wavelet transforms, [INFO.INFO-NI]Computer Science [cs]/Networking and Internet Architecture [cs.NI], 020204 information systems, Histogram, 0202 electrical engineering, electronic engineering, information engineering, Computer vision, Image retrieval, Sparse representation, Mathematics, business.industry, Wavelet transform, Reconstruction error, Pattern recognition, Sparse approximation, Discrete wavelet transforms, Vectors, [SPI.TRON]Engineering Sciences [physics]/Electronics, Feature representation, Graphic methods, Computer Science::Computer Vision and Pattern Recognition, 020201 artificial intelligence & image processing, Artificial intelligence, Color texture retrieval, Retrieval rate, business
Abstract

International audience; Sparse representation is proposed to generate the histogram of feature vectors, namely sparse representation based histogram (SRBH), in which a feature vector is represented by a number of basis vectors instead of by one basis vector in classical histogram. This amelioration makes the SRBH to be a more accurate representation of feature vectors, which is confirmed by the analysis in the aspect of reconstruction errors and the application in color texture retrieval. In color texture retrieval, feature vectors are constructed directly from coefficients of Discrete Wavelet Transform (DWT). Dictionaries for sparse representation are generated by K-means. A set of sparse representation based histograms from different feature vectors is used for image retrieval and chisquared distance is adopted for similarity measure. Experimental results assessed by Precision-Recall and Average Retrieval Rate (ARR) on four widely used natural color texture databases show that this approach is robust to the number of wavelet decomposition levels and outperforms classical histogram and state-of-the-art approaches. © Springer International Publishing AG 2016.

Published
2016
Full Text
View/download PDF

95. Fusion of Textural and Spectral Information for Tree Crop and Other Agricultural Cover Mapping With Very-High Resolution Satellite Images

Author

Camille Lelong, Kidiyo Kpalma, Ahsan Ahmad Ursani, Joseph Ronsin, Institut d'Electronique et de Télécommunications de Rennes (IETR), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Ecole Supérieure d'Electricité - SUPELEC (FRANCE)-Centre National de la Recherche Scientifique (CNRS), Institute of Information and Communication Technologies (IICT), Mehran University of Engineering & Technology, Gestion de l'Eau, Acteurs, Usages (UMR G-EAU), Centre de Coopération Internationale en Recherche Agronomique pour le Développement (Cirad)-Centre international d'études supérieures en sciences agronomiques (Montpellier SupAgro)-AgroParisTech-Institut national de recherche en sciences et technologies pour l'environnement et l'agriculture (IRSTEA)-Institut de Recherche pour le Développement (IRD [France-Sud]), Centre National de la Recherche Scientifique (CNRS)-Ecole Supérieure d'Electricité - SUPELEC (FRANCE)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES), Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), and Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-SUPELEC-Centre National de la Recherche Scientifique (CNRS)

Subjects
Terre agricole, fusion, tree crops, Atmospheric Science, 010504 meteorology & atmospheric sciences, Télédétection, Computer science, Multispectral image, Feature extraction, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, 0211 other engineering and technologies, 02 engineering and technology, 01 natural sciences, land-use mapping, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, Image texture, Computer vision, Computers in Earth Sciences, Image resolution, 021101 geological & geomatics engineering, 0105 earth and related environmental sciences, Cartographie, Contextual image classification, business.industry, Visual comparison, Image segmentation, Remote sensing, Panchromatic film, ComputingMethodologies_PATTERNRECOGNITION, classification, P31 - Levés et cartographie des sols, Computer Science::Computer Vision and Pattern Recognition, Artificial intelligence, U30 - Méthodes de recherche, business, texture, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing
Abstract

International audience; A new procedure is proposed for agricultural land-use mapping that addresses a known weakness of classical per-pixel methods in situations involving mixed tree crops. The proposed scheme uses a pair of very-high resolution satellite-borne panchromatic and multispectral images and integrates classification results of two parallel and independent analyses, respectively based on spectral and textural information. The multispectral image is divided into spectrally homogeneous but non-contiguous segments using unsupervised classification. In parallel, the panchromatic image is split into a grid of square blocks on which is performed a texture-driven supervised classification. Finally, the spectral and the textural classifications are fused to generate the land-use map. This method contrasts with object-based methods that sequentially perform image segmentation and classification. Results are evaluated both quantitatively and qualitatively, based on field survey ground-truth data. The quantitative assessment is presented in terms of overall accuracy (from 80% to 100% depending on the area) and Kappa coefficients. Visual comparison of the resulting map with the ground-truth is performed, with the analysis of the binary error maps. Merging spectral and textural classifications results in finer border delimitation and improves the overall classification accuracy of agricultural land-use by 27% as compared to textural classification alone.

Published
2012
Full Text
View/download PDF

96. Image steganography based on digital holography and saliency map

Author

Lu Zhang, Shuming Jiao, Kidiyo Kpalma, Wenbin Zou, Zhuang Zhaoyong, Shenzhen Univerisity [Shenzhen], Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA), Université de Nantes (UN)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES), and Nantes Université (NU)-Université de Rennes 1 (UR1)

Subjects
Computer science, Image quality, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Holography, 02 engineering and technology, 01 natural sciences, law.invention, 010309 optics, 020210 optoelectronics & photonics, law, 0103 physical sciences, 0202 electrical engineering, electronic engineering, information engineering, Computer vision, Saliency map, Digital watermarking, Image retrieval, ComputingMilieux_MISCELLANEOUS, Steganography, Pixel, business.industry, General Engineering, Digital imaging, Watermark, Atomic and Molecular Physics, and Optics, Artificial intelligence, business, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Host (network), Digital holography
Abstract

A gray-level intensity image can be employed as a host image for hiding a watermark image to protect information security. Past research works demonstrate that a gray-level hidden image can be embedded into the host image by a digital phase-only holography method. However, the fidelity of retrieved watermark image from the host image is not very satisfactory and the host image quality is degraded due to the insertion of external data bits, especially when observers focus on the saliency regions in the host image. To address this problem, we propose a steganography method based on digital holography and the saliency map of the host image. First, we calculate the hidden capacity (number of bits to be replaced) for each host image pixel based on the weighted sum of pixel intensity and saliency value. Next, a multilevel phase-only digital hologram of the watermark image will be calculated by the Gerchberg–Saxton method under the constraint of the hidden capacity of host image. Finally, we embedded the multilevel phase-only digital hologram into the host image by replacing a corresponding number of bits in each pixel. In this way, the host image can preserve good image fidelity for its saliency regions even if we hide a large amount of digital hologram data into the host image. The experimental results show that the quality of retrieved watermark image from the host image and the quality of saliency regions in the watermarked host image, in our proposed scheme, are superior to the state-of-the-art works reported.

Published
2019
Full Text
View/download PDF

97. Parallel algorithm implementation for multi-object tracking and surveillance

Author

Nasreddine Taleb, Mohamed Elbahri, Kidiyo Kpalma, Joseph Ronsin, Université Djilali Liabès [Sidi-Bel-Abbès], Institut d'Électronique et des Technologies du numéRique (IETR), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Institut National des Sciences Appliquées Rouen, United Technologies, Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Université de Nantes (UN)-Université de Rennes 1 (UR1)

Subjects
Computer science, Parallel algorithm, Graphics processing unit, CUDA, 02 engineering and technology, multi-object tracking, surbveillance, 0202 electrical engineering, electronic engineering, information engineering, Computer vision, Parallel OMP, Parallel implementation, Contextual image classification, business.industry, Cognitive neuroscience of visual object recognition, 020206 networking & telecommunications, Pattern recognition, Sparse approximation, Matching pursuit, Feature (computer vision), Video tracking, 020201 artificial intelligence & image processing, Computer Vision and Pattern Recognition, Artificial intelligence, business, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Software
Abstract

International audience; A recently developed sparse representation algorithm, has been proved to be useful for multi-object tracking and this study is a proposal for developing its parallelisation. An online dictionary learning is used for object recognition. After detection, each moving object is represented by a descriptor containing its appearance features and its position feature. Any detected object is classified and indexed according to the sparse solution obtained by an orthogonal matching pursuit (OMP) algorithm. For a real-time tracking, the visual information needs to be processed very fast without reducing the results accuracy. However, both the large size of the descriptor and the growth of the dictionary after each detection, slow down the system process. In this work, a novel accelerating OMP algorithm implementation on a graphics processing unit is proposed. Experimental results demonstrate the efficiency of the parallel implementation of the used algorithm by significantly reducing the computation time.

Published
2016
Full Text
View/download PDF

98. Region-based image retrieval in the compressed domain using shape-adaptive DCT

Author

Kamel Belloulata, Kidiyo Kpalma, Amina Belalia, Université Djilali Liabès [Sidi-Bel-Abbès], Institut d'Électronique et des Technologies du numéRique (IETR), Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Université de Nantes (UN)-Université de Rennes 1 (UR1), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), and Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)

Subjects
Computer Networks and Communications, Computer science, Feature vector, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Image processing, 02 engineering and technology, Similarity measure, Digital image, Segmentation, Image texture, Region-Based Image Retrieval (RBIR), Histogram, 0202 electrical engineering, electronic engineering, information engineering, Media Technology, Discrete cosine transform, Computer vision, Visual Word, Content-Based Image Retrieval (CBIR), Image retrieval, SA-DCT, ComputingMilieux_MISCELLANEOUS, Feature detection (computer vision), business.industry, Quantization (signal processing), DCT, 020207 software engineering, Pattern recognition, computer.file_format, Semantic image retrieval, JPEG, Automatic image annotation, Hardware and Architecture, Feature (computer vision), 020201 artificial intelligence & image processing, Artificial intelligence, business, computer, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Software
Abstract

Content-based image retrieval (CBIR) has drawn substantial research and many traditional CBIR systems search digital images in a large database based on features, such as color, texture and shape of a given query image. A majority of images are stored in compressed format and most of compression technologies adopt different kinds of transforms to achieve compression. Therefore, features can be extracted directly from images in compressed format by using, for example, discrete cosine transform (DCT) for JPEG compressed images. Region-based image retrieval (RBIR) is an image retrieval approach which focuses on contents from regions of images, instead of the content from the entire image in early CBIR. Although RBIR approaches attempt to solve the semantic gap problem existed in global low-level features in CBIR by using local low-level features based on regions of images. This paper proposes a new RBIR approach using Shape adaptive discrete cosine transform (SA-DCT). At a bottom level, local features are constructed from the coefficients of quantized block transforms (low-level features) for each region. Quantization acts for the concentration of block-wise information in a more condense way, which is highly desirable for the retrieval tasks. At an intermediate level, histograms of local image features are used as descriptors of statistical information. Finally, at the top level, the combination of histograms from different image regions (objects) is defined as a way to incorporate high-level semantic information. In this retrieval system, an image has a prior segmentation alpha plane, which is defined exactly as in MPEG-4. Therefore, an image is represented by segmented regions, each of which is associated with a feature vector derived from DCT and SA-DCT coefficients. Users can select any region as the main theme of the query image. The similarity between a query image and any database image is ranked according to a same similarity measure computed from the selected regions between two images. For those images without distinctive objects and scenes, users can still select the whole image as the query condition. The experimental results show that the proposed approach is able to identify main objects and reduce the influence of background in the image, and thus improve the performance of image retrieval in comparison with a conventional CBIR based on DCT.

Published
2016
Full Text
View/download PDF

99. Unsupervised Joint Salient Region Detection and Object Segmentation

Author

Yong Zhao, Kidiyo Kpalma, Joseph Ronsin, Wenbin Zou, Nikos Komodakis, Zhi Liu, College of Information Engineering [Shenzhen], Shenzhen Univerisity [Shenzhen], School of Communication and Information Engineering [Shanghai], Shanghai University, Analysis representation, compression and communication of visual data (Sirocco ), SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE (IRISA-D5), Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-Télécom Bretagne-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes 1 (UR1), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-Télécom Bretagne-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-Télécom Bretagne-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Inria Rennes – Bretagne Atlantique, Institut National de Recherche en Informatique et en Automatique (Inria), Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes 1 (UR1), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Peking University [Shenzhen], Laboratoire d'Informatique Gaspard-Monge (LIGM), Centre National de la Recherche Scientifique (CNRS)-Fédération de Recherche Bézout-ESIEE Paris-École des Ponts ParisTech (ENPC)-Université Paris-Est Marne-la-Vallée (UPEM), imagine [Marne-la-Vallée], Centre National de la Recherche Scientifique (CNRS)-Fédération de Recherche Bézout-ESIEE Paris-École des Ponts ParisTech (ENPC)-Université Paris-Est Marne-la-Vallée (UPEM)-Centre National de la Recherche Scientifique (CNRS)-Fédération de Recherche Bézout-ESIEE Paris-École des Ponts ParisTech (ENPC)-Université Paris-Est Marne-la-Vallée (UPEM)-Centre Scientifique et Technique du Bâtiment (CSTB), 61401287, National Natural Science Foundation of China, 20134408110001, Chinese Specialized Research Fund for the Doctoral Program of Higher Education, CXB201105060068, Shenzhen Key Laboratory Project, FP7-ICT-611145, European Commission through the ROBOSPECT Project, Inria Rennes – Bretagne Atlantique, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE (IRISA-D5), CentraleSupélec-Télécom Bretagne-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National de Recherche en Informatique et en Automatique (Inria)-École normale supérieure - Rennes (ENS Rennes)-Université de Bretagne Sud (UBS)-Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Télécom Bretagne-Université de Rennes 1 (UR1), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-École normale supérieure - Rennes (ENS Rennes)-Université de Bretagne Sud (UBS)-Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA), Centre National de la Recherche Scientifique (CNRS)-CentraleSupélec-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Université de Nantes (UN)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-Inria Rennes – Bretagne Atlantique, Nantes Université (NU)-Université de Rennes 1 (UR1), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-Télécom Bretagne-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Institut National de Recherche en Informatique et en Automatique (Inria)-Télécom Bretagne-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Institut de Recherche en Informatique et Systèmes Aléatoires (IRISA), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Université de Bretagne Sud (UBS)-École normale supérieure - Rennes (ENS Rennes)-Télécom Bretagne-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Université Paris-Est Marne-la-Vallée (UPEM)-École des Ponts ParisTech (ENPC)-ESIEE Paris-Fédération de Recherche Bézout-Centre National de la Recherche Scientifique (CNRS), and Université Paris-Est Marne-la-Vallée (UPEM)-École des Ponts ParisTech (ENPC)-ESIEE Paris-Fédération de Recherche Bézout-Centre National de la Recherche Scientifique (CNRS)-Université Paris-Est Marne-la-Vallée (UPEM)-École des Ponts ParisTech (ENPC)-ESIEE Paris-Fédération de Recherche Bézout-Centre National de la Recherche Scientifique (CNRS)-Centre Scientifique et Technique du Bâtiment (CSTB)

Subjects
Markov random field, Boosting (machine learning), Saliency detection, business.industry, Segmentation-based object categorization, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Scale-space segmentation, Pattern recognition, Image segmentation, object segmentation, Computer Graphics and Computer-Aided Design, Region growing, Computer vision, Segmentation, Artificial intelligence, business, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Software, low-rank matrix recovery, Sparse matrix, Mathematics
Abstract

International audience; This paper presents a novel unsupervised algorithm to detect salient regions and to segment out foreground objects from background. In contrast to previous unidirectional saliency-based object segmentation methods, in which only the detected saliency map is used to guide the object segmentation, our algorithm mutually exploits detection/segmentation cues from each other. To achieve this goal, an initial saliency map is generated by the proposed segmentation driven low-rank matrix recovery model. Such a saliency map is exploited to initialize object segmentation model, which is formulated as energy minimization of Markov random field. Mutually, the quality of saliency map is further improved by the segmentation result, and serves as a new guidance for the object segmentation. The optimal saliency map and the final segmentation are achieved by jointly optimizing the defined objective functions. Extensive evaluations on MSRA-B and PASCAL-1500 datasets demonstrate that the proposed algorithm achieves the state-of-the-art performance for both the salient region detection and the object segmentation.

Published
2015
Full Text
View/download PDF

100. Compressed Sensing Performance of Random Bernoulli Matrices with High Compression Ratio

Author

Weiyu Li, Weizhi Lu, Kidiyo Kpalma, Joseph Ronsin, Institut d'Électronique et des Technologies du numéRique (IETR), Université de Nantes (UN)-Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), Centre de Recherche en Economie et Statistique [Bruz] (CREST), Ecole Nationale de la Statistique et de l'Analyse de l'Information [Bruz] (ENSAI), Université de Nantes (UN)-Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS), and Nantes Université (NU)-Université de Rennes 1 (UR1)

Subjects
Discrete mathematics, Mathematical optimization, high dimension, Applied Mathematics, compression ratio, Bernoulli distribution, random matrix, Electronic mail, Restricted isometry property, Bernoulli's principle, Compressed sensing, Signal Processing, Compression ratio, Electrical and Electronic Engineering, Random matrix, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Sparse matrix, Mathematics, compressed sensing
Abstract

This letter studies the sensing performance of random Bernoulli matrices with column size $n$ much larger than row size $m$ . It is observed that as the compression ratio $n/m$ increases, this kind of matrices tends to present a performance floor regarding the guaranteed signal sparsity. The performance floor is effectively estimated with the formula ${1 \over 2}(\sqrt {\pi m/2} + 1)$ . To the best of our knowledge, it is the first time in compressed sensing, a theoretical estimation is successfully proposed to reflect the real performance.

Published
2015
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results